Mark Wiggins Contents ALPHA-X project W hat is a L W FA? - - PowerPoint PPT Presentation

Laser W akefield Accelerator (LW FA): towards a compact light source The

Mark Wiggins

Contents

• ALPHA-X project
• W hat is a L

W FA?

• Motivation:

quality electron beams and light sources

• The ALPHA-X beam line:

experimental setup

• Experimental results:

pointing and energy stability, charge, energy spread, emittance, bunch length

• L

W FA and beam transport simulations

• O utlook for free-electron laser (FEL) driven by L

W FA beam

• Summary

Advanced Laser Plasma High-energyAccelerators towards X-rays

• BasicTechnology grant (2002) and EPSRC grant (2007)
• Consortium of U.K. research teams (Stage 2)

U.

• St. Andrews
• A. Cairns

U. Dundee

• A. Gillespie

U. Strathclyde

• D. Jaroszynski
• B. Bingham
• K. Ledingham
• P. McKenna

Cockcroft Institute

• M. Poole
• R. Tucker

Partners – L. Silva & T. Mendonca (IST), B. Cros (UPS - LPGP), W. Leemans (LBNL),

• B. van der Geer & M. de Loos (Pulsar Phys), G. Shvets (UTA), J. Zhang (CAS)

U. Abertay Dundee

• A. MacLeod

And numerous collaborators

ALPHA- X Project

Group Leader: Prof. Dino Jaroszynski Experiments: Riju Issac, Gregor Welsh, Enrico Brunetti, Gregory Vieux PhDs: Richard Shanks, Maria Pia Anania, Silvia Cipiccia, Salima Abuazoum, Grace Manahan, Constantin Aniculaesei, Anna Subiel, David Grant Theory: Bernhard Ersfeld, Ranaul Islam, Gaurav Raj, Adam Noble PhDs: John Farmer, Sijia Chen, Ronan Burgess, Yevgen Kravets Technicians: David Clark, Tom McCanny Visiting Professor: Rodolfo Bonifacio

ALPHA- X Project

Scottish Universities Physics Alliance

The LWFA

• Tajima & Dawson PRL 43, 267 (1979).
• Intense femtosecond laser propagating

in underdense plasma.

• Relativistically self-guided channel.
• Ponderomotive force leads to charge

separation and plasma density wake.

• Electrons trapped at back of bubble

and accelerated in the very large electrostatic fields.

• Electron velocity (~c) > laser group velocity and electrons catch up on laser.
• Energy at dephasing length:

laser self-injected electron bunch undergoing betatron oscillations ion bubble

Motivation

User Facilities: SSRL synchrotron LCLS X-ray FEL RF Linac: 3.2 km long 50 GeV electrons

16 MeV/m gradient

• Conventional synchrotrons and FELs are very large
• A L

W FA-driven light source is ultra-compact

• Accelerating gradient ~100 GeV/m
• Great uses:

short pulses, small source sizes

• W ider accessibility

ALPHA-X Length ~10 m

Conventional v Plasma Accelerators

Plasma waves RF Cavities

• Max. E field ~100 MV/m
• Limited by breakdown
• 1000 times smaller & cheaper
• 1 GeV in 33 mm capillary (LBN L 2006)

Strathclyde Capillary

Our goal

L W FAs to date

• High charge density:

10’s of pC in inferred ~ 10 fs (peak current I ~ kA)

• Low emittance:

inferred εN ~ few π mm mrad (no direct measurements)

• Significant relative energy spread σγ/γ ~ 1 – 2% at best
• X-ray FEL needs σγ/γ ~0.1%
• We are looking to produce high quality electron beams (high I, low εN , low σγ/γ)
• And to apply them in useful ways:
• Medical imaging
• Ultrafast probing
• Detector development for nuclear physics
• Strathclyde/Glasgow/Institute for Cancer Research project (e− beam therapy)
• Future plans at the end...

Synchrotron / undulator radiation

• Relativistic electrons in a magnetic field follow a curved trajectory and

i.e. they are accelerated.

• Radiation emitted into a narrow cone (lab frame of reference).
• Single magnet:

synchrotron, Magnet array: undulator or wiggler.

N

cen

γ ≅ θ 1 N 1 = λ λ Δ

• Undulator Equation

where h is the harmonic order and K = λueB/2πm0c < 1 period λu N periods

Undulator

⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ γ θ + + γ λ = λ

2 2 2 2

2 1 2 K h

u

LWFA undulator radiation

• J

ena / Strathclyde / Stellenbosch experiment

• 55-70 MeV electrons
• VIS/IR synchrotron radiation

Schlenvoigt et al., N ature Phys. 4, 130 (2008) Gallacher et al.,

• Phys. Plasmas 16, 093102 (2009)

LWFA undulator radiation

• MPQ / FZD / O xford experiment
• 150-210 MeV electrons
• XUV synchrotron radiation

Fuchs et al., N ature Phys. 5, 826 (2009)

• N ext step:

Free-electron laser for 106 – 108 increase in photon output

• High FEL gain criteria:

εn < λγ/4π and σγ/γ < ρ

• N eed the beam quality and good transport...

ALPHA- X Beam Line

• Laser:

λ0 = 800 nm, E = 900 mJ , τ = 35 fs, P = 26 TW, I = 2 × 1018 W cm-2, initial a0 = 1.0

• Gas J

et: helium, 2 mm nozzle, ne ≈ 1 – 5 × 1019 cm-3

• Q uadrupole magnets:

permanent (PMQ s) & electromagnetic (EMQ s)

• Beam profile monitors:

pop-in Lanex screens / Ce:YAG crystals

• Diagnostics:

pop-in emittance mask & pop-in aluminium pellicle for transition radiation

Accelerator Pepper pot PMQs EMQs Electron Spectrometer Undulator Pellicle

Electron Spectrometer

Dual function device High resolution chamber Resolution – design ~ 0.1% Electron energy up to 105 MeV (Bmax = 1.65 T) High energy chamber Uses upstream quadrupoles to aid focusing Energy resolution ~0.2 – 10% (energy dependent) Electron energy up to ~ 660 MeV (Bmax = 1.65 T) Ce:YAG crystal 300 × 10 × 1 mm 14-bit PGR Grasshopper camera not shown

• Designed by Allan Gillespie / Allan MacLeod
• Built by Sigmaphi (France)

Experimental Results – beam pointing

• 500 consecutive shots
• narrow divergence (~2 mrad) beam
• wide divergence halo
• θX = (7 ± 3) mrad, θY = (3 ± 2) mrad

5 mrad

• 8 mrad acceptance angle for EMQ s
• 25% pointing reduction with

PMQ s installed

no PMQs PMQs in

Experimental Results – PMQs

• 1.5 T magnets (similar to the MPQ design)
• Triplet settings for collimation of the “main peak” monoenergetic electron bunch
• Swirls due to low energy halo electrons

PMQs in no PMQs

Experimental Results – energy stability

Electron Spectrometer: 200 consecutive shots (spectrum on 196 shots) 69 90 124 185

Energy (MeV)

69 90 124 185

Energy (MeV)

100 consecutive shots Mean E0 = (137 ± 4) MeV 2.8% stability

Experimental Results – charge

LANEX 2 Imaging Plate

All screens now calibrated

Experimental Results – energy spectra I

NO QUADS QUADS QUADS NO QUADS

i.e. to measure small spreads, emittance must be small!

electron beam energy = 83 MeV r.m.s. source size = 2 μm spectrometer field = 0.59 T zero energy spread electron beam energy = 83 MeV r.m.s. source size = 2 μm spectrometer field = 0.59 T emittance εN = 0.5π mm mrad zero energy spread

Simulations of electron spectrometer response

NO QUADS QUADS

• General Particle Tracer (GPT) code
• Analytical B field (fringe field responsible for the butterfly profile at 0% spread)

Experimental Results – energy spectra I I

• Scaling of central energy and energy spread with charge

Beam loading Beam loading

• W iggins et al., PPCF 52, 124032 (2010).

Experimental Results – energy spectra I I I

σγ/γ MEAS = 0.7%

simulation at 146 MeV

σγ/γ MEAS = 0.4%

simulation at 85 MeV

Experimental Results – energy spectra I I I

• 2mm gas jet:

accelerating gradient ≈1 GeV/cm

• A hint of a fixed absolute energy spread ~ 0.6-0.8 MeV

E0 = 172 MeV

• meas. σE = 1.3 MeV
• meas. σγ / γ = 0.75%

E0 = 210 MeV

Experimental Results – transverse emittance

50000 100000

• 1.0
• 0.5

0.0 0.5 1.0

• 3
• 2
• 1

1 2 3

counts

(b) x' [mrad] x [mm]

θx σx x σx x x′

<x> ∝ I*x - averaged <x’> ∝ I*(θx+ σx) – averaged Emittance (rms):

εx, rms = [<x2> <x’2> - <xx’>2]1/2

Direct Calculation: (Zhang FERMILAB-TM-1988)

• divergence 4 mrad
• hole size correction
• limited by detection system
• εN < (5.5 ± 1)π mm mrad
• Pepper pot mask technique
• First generation mask with hole φ ~ 55 μm

Experimental Results – transverse emittance

• divergence 2-4 mrad for this run

with 125 MeV electrons

• average εN = (2.0 ± 0.6)π mm mrad
• best εN = (1.0 ± 0.1)π mm mrad
• Elliptical beam:

εN , X > εN , Y

• Resolution limited
• Second generation mask with hole φ ~ 25 μm and improved detection system

False colour image of an electron beam with and without the pepper‐pot mask.

Experimental Results – transverse emittance

• Measured emittance consistent

with ~1 fs bunch

• θ ∝ Q1/2 scaling:

implies constant σz

• θ ∝ Q1/3 scaling:

very slow increase

• f σz with Q
• Brunetti et al., Phys. Rev. Lett. 105, 215007 (2010).
• Experiments with third generation mask in progress.

State of play

• Measured low σγ/γ < 1% (→ 0% with spectrometer response)
• Measured εN = 1π mm mrad (detector-limited, inferred ~0.5 π mm mrad)
• Measured στ = 2 fs
• Measured charge Q = 1-5 pC
• W hy do we get these high quality beams?
• O perating in a near-threshold, low charge regime.
• Use PIC simulations and reduced models to understand our accelerator.
• Injection of electrons from a small volume of phase-space.
• Reduced model in progress.

PI C simulations of our LWFA

Measured beam profile Phase-space distribution

Beam loading simulations

• 2-D reduced model
• N o self-injection

(external 6 MeV beam is input)

• O ptimal charge for

flattening potential along beam and obtaining minimum spread No beam loading With beam loading With beam loading and 10 pC change

• λp = 7 μm
• lbunch = 1 μm
• Beam loading reduces the variation in accelerating potential along the bunch

Electron energy (MeV) Radiation λ (nm) Emittance criterion (π mm mrad) Gain parameter ρ Relative energy spread

90 261 3 0.011 0.007 150 94 2 0.006 0.004 500 8 0.6 0.002 0.001(?)

Viability of LWFA- driven FEL

• High FEL gain criteria:

εn < λγ/4π & σγ/γ < ρ

• Experimental εn ≤ 1π mm mrad & σγ/γ ≤ 0.007
• For fixed σγ = 0.6 MeV, σγ/γ reduces at short λ

3 / 1 2

2 2 1 ⎥ ⎥ ⎦ ⎤ ⎢ ⎢ ⎣ ⎡ ⎟ ⎟ ⎠ ⎞ ⎜ ⎜ ⎝ ⎛ =

x u u A p

a I I πσ λ γ ρ

λu = 15 mm, N = 200, au = 0.38

ALPHA-X Undulator

• Actually, need to consider the slice parameters:
• slice εn & σγ/γ in a co-operation length

FEL Simulation

SIMPLEX CO DE SIMULATIO N RESUL TS (100 MeV electrons) Saturation power(1st harmonic): 20 GW @ saturation distance:1.8 m

Synchrotron: Peak Brilliance B = 3 x 1025 photons/sec/mrad2/mm2/0.1% BW Average brilliance B = 2.5 x 1011 for PRF 10 Hz With laser improvements: PRF 1 kHz → average brilliance B >1013 FEL: B >106 times higher

350 400 450 0.0 5.0x10

1.0x10

1.5x10

Photon Flux [Photons/0.1% BW] Photon energy [eV] Peak Brilliance: 2.97 x 1025 photons/sec/mrad²/mm²/0.1%b.w. Photon flux into 200 μrad 350 400 450 1x10

2x10

3x10

Photon Flux [Photons/0.1% BW] Photon energy [eV] SASE FEL

synchrotron radiation matched beam SASE FEL

Strathclyde capillary beams

• RAL Astra Gemini experiment (X-ray betatron radiation)
• 40 mm, 280 μm capillary
• Stable electron beam generation with large plasma discharge time window.

ALPHA- X Summary

• High quality 70 – 180 MeV electron beams produced on the ALPHA-X beam line.
• energy spread, emittance, bunch length and charge are inter-connected.
• low charge for good quality with kA peak current.
• FEL gain should be observable in VUV – XUV spectral range.

Progress is advancing nicely towards a working compact soft X-ray FEL driven by a L W FA electron beam

→ long gas jet, gas cell or capillary accelerator

Funded by

Thank you